Advertisement

Neurofuzzy-Chaos Engineering for Building Intelligent Adaptive Information Systems

  • Nikola K. Kasabov
  • Robert Kozma

Abstract

Intelligent adaptive information systems are systems which can automatically adapt their structure and behaviour in order to react better to a dynamically changing environment, and to provide knowledge which explains it. Several hybrid fuzzyneuro techniques have already proved to be very useful for this purpose, one of them being the fuzzy neural networks. Fuzzy neural networks have important and useful features, such as: adaptive learning, good generalisation, good explanation facilities in form of fuzzy rules, abilities to accommodate both data and existing knowledge about the problem, ability to act autonomously in a dynamically changing environment. In order to design and train a fuzzy neural network for a particular task in a dynamically changing environment, one need to carefully investigate the type of the dynamics, and the level of chaos in the analysed process. This chapter introduces a way of using both chaos theory and a particular fuzzy neural network, called FuNN, for building adaptive, intelligent multimodular systems. A properly designed and trained FuNN can structurally capture major characteristics of a complex process under control. The use of this methodology for building intelligent adaptive systems is illustrated through examples from control and prediction.

Keywords

Fractal Dimension Membership Function Fuzzy Rule Flow Signal Connection Weight 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. [1]
    N. Kasabov, Foundations of Neural Networks, Fuzzy Systems and Knowledge Engineering, The MIT Press, CA, MA (1996).zbMATHGoogle Scholar
  2. [2]
    T. Yamakawa, H. Kusanagi, E. Uchino and T. Mild, “A new Effective Algorithm for Neo Fuzzy Neuron Model,” Proc. Fifth IFSA World Congress, 1017–1020 (1993).Google Scholar
  3. [3]
    T. Hashiyama, T. Furuhashi, Y. Uchikawa, “A Decision Making Model Using a Fuzzy Neural Network,” Proc. 2nd Int. Conf on Fuzzy Logic & Neural Networks, Iizuka, Japan, 1057–1060 (1992).Google Scholar
  4. [4]
    N. Kasabov, “Learning fuzzy rules and approximate reasoning in neuro-fuzzy hybrid systems,” Fuzzy Sets and Systems (1996).Google Scholar
  5. [5]
    T. Furuhashi, Hasegawa, T., Horikawa S., Uchikawa, Y., “An Adaptive Fuzzy Controller Using Fuzzy Neural Networks,” Proc. Fifth IFSA World Congress, 769–772 (1993).Google Scholar
  6. [6]
    M. M. Gupta, D. H. Rao, “On the principles of fuzzy neural networks,” Fuzzy Sets and Systems 61:1, 1–18 (1994).MathSciNetCrossRefGoogle Scholar
  7. [7]
    M. Brown and C. Harris, Neurofuzzy Adaptive Modelling and Control, Prentice Hall (1994).Google Scholar
  8. [8]
    N. Kasabov, “Adaptable Neuro Production Systems,” Neurocomputing 13, 95–117 (1996).CrossRefGoogle Scholar
  9. [9]
    N. Kasabov, “Investigating the adaptation and forgetting in fuzzy neural networks by using the method of training and zeroing,” Proc. ICNN’96, Plenary, Panel and Special Sessions, 118–123 (1996).Google Scholar
  10. [10]
    N. Kasabov, “Hybrid Connectionist Fuzzy Production Systems—Towards Building Comprehensive AI”, Intelligent Automation and Soft Computing 1:4, 351–360 (1995).Google Scholar
  11. [11]
    N. Kasabov et al., “FuNN—A fuzzy neural network architecture for adaptive learning and knowledge acquisition in multimodular distributed environments,” Information Sciences: Applications, Prentice Hall, 1997, to be published.Google Scholar
  12. [12]
    N. Kasabov, “Adaptive Learning in Modular Fuzzy Neural Networks,” Proc. Int. Conf. on Neuro Information Processing ICONIP’96, Springer Verlag, 1096–1102 (1996).Google Scholar
  13. [13]
    M. Sakuma, R. Kozma, M. Kitamura, “Detection and Characterisation of Anomalies by Applying Methods of Fractal Analysis,” Nucl. Technol. 113, 86–99 (1996).Google Scholar
  14. [14]
    R. Kozma, N. K. Kasabov, T. Cohen, “Integrating Methods of Chaotic Time Series Analysis and Prediction of Process Data in a Hybrid Connectionist Based Environment,” to be published.Google Scholar
  15. [15]
    T. Higuchi, “Approach to an Irregular Time Series on the Basis of the Fractal Theory,” Physica D 31, 277 (1988).MathSciNetzbMATHCrossRefGoogle Scholar
  16. [16]
    T. Higuchi, “Relationship between the Fractal Dimension and the Power Law Index for a Time Series: Num. Investig.,” Physica D 46, 254 (1990).zbMATHCrossRefGoogle Scholar
  17. [17]
    H. Bai-Lin, Chaos II, World Scientific (1990).Google Scholar
  18. [18]
    M. J. Embrechts, Y. Danon, “Determining the fractal dimension of a time series with a neural net,” in: Intelligent engineering systems through artificial neural networks, ed. C. H. Dagli et al., ASME Press, NY, Vol. 3, 897–902 (1993).Google Scholar
  19. [19]
    R. Reed, “Pruning Algorithms—A Survey,” IEEE Tr. Neur. Netw. 4:5, 740–747 (1993).CrossRefGoogle Scholar
  20. [20]
    M. Ishikawa, “Structural Learning with Forgetting,” IEEE Tr. Neur. Netw. 9, 509–521 (1996).CrossRefGoogle Scholar
  21. [21]
    R. Kozma, M. Sakuma, Y. Yokoyama, M. Kitamura, “On the accuracy of mapping by neural networks trained by backpropagation with forgetting,” Neurocomputing 13, 295–311 (1996).CrossRefGoogle Scholar
  22. [22]
    A. Cohen et al., “Application of Computational Intelligence for On-line Control of a Sequencing Batch Reactor at Morrinsville Sewage Treatment Plant,” Proc. IAWQ Conf. Advance Wastewater Treatment, 22–27, The Netherlands, September 1996.Google Scholar

Copyright information

© Springer Science+Business Media New York 1997

Authors and Affiliations

  • Nikola K. Kasabov
    • 1
  • Robert Kozma
    • 1
  1. 1.Dept of Information ScienceUniversity of OtagoDunedinNew Zealand

Personalised recommendations